EEVblog Electronics Community Forum
Products => Test Equipment => Topic started by: Neodymium on December 26, 2016, 07:22:00 pm
-
Hello Everyone,
I've many test gears inside my own lab. Rather than using isolation transformer with has the price almost the same as UPS
I would like to ask if it is possible to isolate each equipment using individual small UPS like this cheap one
http://www.apcbyquickserv.com/products/category/detail/index.php?idss=1900 (http://www.apcbyquickserv.com/products/category/detail/index.php?idss=1900)
and then operate each gears using offline mode base on UPS battery. Does this approach give a true isolation from main?
This APC unit has step sine wave output but not pure sine wave, I would like to know pros. and cons. from all of you guys.
Thanks, ;D
Neodymium
-
There are many cheap ways to provide isolation for equipment......not all of them safe.
First cost = last cost and is the safest:
Pintek DP-25 or similar Differential probe.
-
Thanks for your advise tautech,
I've already had differential probe, mine is Hameg HZ100. But I need to isolate all of 4 channels from main which required
at least isolation transformer and that diff-probe cost much more than UPS or iso-transformer anyway.
So I would like to know have you guys ever try to set up this approach succeed?
Thank alot,
Neodymium
-
Thanks for your advise tautech,
I've already had differential probe, mine is Hameg HZ100. But I need to isolate all of 4 channels from main which required
at least isolation transformer and that diff-probe cost much more than UPS or iso-transformer anyway.
So I would like to know have you guys ever try to set up this approach succeed?
Thank alot,
Neodymium
WE NEVER know the experience or knowledge of members like you when they ask such questions, but generally just asking demonstrates a lack of understanding of the risks.
You already know about Differential probes so we know you're just trying to do this on the cheap.
Cheap is not always safe, buy extra probes and sell them when your project is done.
-
An UPS is not suitable for that purpose because its output is not isolated from ground/earth/metal chassis. And if you add the costs for battery replacements a proper isolation transformer or differential probes are less expensive on the long term.
-
An UPS is not suitable for that purpose because its output is not isolated from ground/earth/metal chassis. And if you add the costs for battery replacements a proper isolation transformer or differential probes are less expensive on the long term.
I suppose the OP didn't state this, but obviously one must disconnect the UPS from mains (e.g. operate in battery powered mode) in order to isolate things.
This is a fine technique. Virtually no UPSes have any exposed metal contacts whatsoever, so there is no additional risk of human-touch contact.
-
Please have a look at an APC Smart-UPS and its circuit diagram.
-
Virtually no UPSes have any exposed metal contacts whatsoever, so there is no additional risk of human-touch contact.
Excepting the scope metal chassis is still commoned to the BNC outer shell.
Live scope. :scared:
-
I suppose the OP didn't state this, but obviously one must disconnect the UPS from mains (e.g. operate in battery powered mode) in order to isolate things.
This is a fine technique. Virtually no UPSes have any exposed metal contacts whatsoever, so there is no additional risk of human-touch contact.
The last two UPS systems I tried did not power on without plugging them into the (dead) outlet. It is a safety feature I assume. If so then you can't get any isolation.
-
All the isolation tricks can offer is only one truly isolated channel, if you want/need 2 or more isolated channels to connect to multiple (at different potentials) reference points then an isolated channel scope or Differential probes are the only options.
-
Interestingly, my philips scope came with a 2 prong plug, and can also be operated on 24V DC. If you're looking for an isolated scope, have a look at finding an old PM34xx
-
Thanks for your advise tautech,
I've already had differential probe, mine is Hameg HZ100. But I need to isolate all of 4 channels from main which required
at least isolation transformer and that diff-probe cost much more than UPS or iso-transformer anyway.
So I would like to know have you guys ever try to set up this approach succeed?
Thank alot,
Neodymium
WE NEVER know the experience or knowledge of members like you when they ask such questions, but generally just asking demonstrates a lack of understanding of the risks.
You already know about Differential probes so we know you're just trying to do this on the cheap.
Cheap is not always safe, buy extra probes and sell them when your project is done.
Hello tautech,
I'm a chief R&D engineer responsible in development of power line networking devices base on BPSK, OFDM and
DCSK (Differential Code Shift Keying) technologies which provide data throughput of at least 7kbps up to 100Mbps
through a noisy power line environment for more than 10 year.
Does this information MAKE MYSELF CLEAR enough for my knowledge against yours?
I pretty understand the risk of electrical isolation as from my job experience. But do you understand the basic topology
and concept of such kind of isolation transformer and UPS equipment? A galvanic isolation of transformer inside both devices
are a nearly perfect isolation compare to the isolation base on a high impedance in such differential probe.
A CHEAP APPROACH IS NOT ALWAYS RISK, it depends on how much knowledge you have to tinkering things base on a
correct electrical principle.
If you are not willing to give contribution, just let the other people that willing to share an experience to find an alternative
ways to solve this requirement in CHEAP solution instead. Thank for your cooperation.
Neodymium
-
Hello tautech,
I'm a chief R&D engineer responsible in development of power line networking devices base on BPSK, OFDM and
DCSK (Differential Code Shift Keying) technologies which provide data throughput of at least 7kbps up to 100Mbps
through a noisy power line environment for more than 10 year.
Does this information MAKE MYSELF CLEAR enough for my knowledge against yours?
I pretty understand the risk of electrical isolation as from my job experience. But do you understand the basic topology
and concept of such kind of isolation transformer and UPS equipment? A galvanic isolation of transformer inside both devices
are a nearly perfect isolation compare to the isolation base on a high impedance in such differential probe.
A CHEAP APPROACH IS NOT ALWAYS RISK, it depends on how much knowledge you have to tinkering things base on a
correct electrical principle.
If you are not willing to give contribution, just let the other people that willing to share an experience to find an alternative
ways to solve this requirement in CHEAP solution instead. Thank for your cooperation.
Neodymium
You now have moved the goalposts. :)
From what you have indicated thus far you only need to remove the mains ground reference and not need 4 truly isolated channels, correct ?
Another thing you have not shared is the voltages you'll be connecting to, hence my generic replies in order to keep all that are watching and those that may find this thread SAFE as many do not have the experience you do.
At low levels of voltage exposure then you could just float your scope, a method I'd not normally suggest but maybe it's the simplest solution for you. Generally frowned on of course. ;)
-
All the isolation tricks can offer is only one truly isolated channel, if you want/need 2 or more isolated channels to connect to multiple (at different potentials) reference points then an isolated channel scope or Differential probes are the only options.
This is the best answer in this thread so far. It sounds like the OP was thinking that if he got 4x UPSes or 4x isolation transformers that he could isolate the 4 ground leads inside the oscilloscope from each other. This is not possible with the vast majority of scopes. Typically some battery powered scopes will have each channel isolated from each other, and then some bench and/or higher end scopes might.
In other words, regardless of your isolation scheme, once you connect your first ground clip into a circuit -- that is effectively connected to the outer BNC (or whatever) connector for all channels.
-
ways to solve this requirement in CHEAP solution instead. Thank for your cooperation.
CHEAP and SAFE don't go together. I feel sorry for your employees having to work in an unsafe environment. The lawyer hired by their relatives to sue your employer due to death caused by negligence will have a good time though when he finds outs you sought advice and choose to ignore it to cut costs and make you look good to the people upstairs because you didn't go over budget. How happy will they be when someone gets injured or (worse) killed on your watch and who will they blame?
If you want to do isolated measurements then use an oscilloscope with CAT rated isolated inputs (there are several on the market ranging from reasonable to excellent) or use differential probes with the appropriate CAT rating.
-
I'm a chief R&D engineer responsible in development of power line networking devices base on BPSK, OFDM and
DCSK (Differential Code Shift Keying) technologies which provide data throughput of at least 7kbps up to 100Mbps
through a noisy power line environment for more than 10 year.
Does this information MAKE MYSELF CLEAR enough for my knowledge against yours?
No.
The most it indicates is what you think you know.
-
And someday someone simply forgets to pull the plug... :palm:
-
When I firstly saw the name of the topic I thought that the OP wanted to measure some node referenced to another floating node on some low voltage circuit. :o
Anyway, it is possible to do it ?
-
ways to solve this requirement in CHEAP solution instead. Thank for your cooperation.
CHEAP and SAFE don't go together. I feel sorry for your employees having to work in an unsafe environment. The lawyer hired by their relatives to sue your employer due to death caused by negligence will have a good time though when he finds outs you sought advice and choose to ignore it to cut costs and make you look good to the people upstairs because you didn't go over budget. How happy will they be when someone gets injured or (worse) killed on your watch and who will they blame?
If you want to do isolated measurements then use an oscilloscope with CAT rated isolated inputs (there are several on the market ranging from reasonable to excellent) or use differential probes with the appropriate CAT rating.
Just so.
Too many "engineers" only think of the happy-days everything-working-optimally scenario. "Real" engineers think about how things will fail, and how to avoid failures. Some of the failure modes with mains (and higher) voltages are surprising, subtle, rare - and lethal. Not a good combination, and fertile ground for the Dunning-Kruger effect :(
Personally I don't care what people do to their own bodies, and don't mind paying for their medical bills. I do object to someone killing (or worse seriously injuring) other people.
-
ways to solve this requirement in CHEAP solution instead. Thank for your cooperation.
The lawyer hired by their relatives to sue your employer due to death caused by negligence
Those kind of accidents are investigated by the health and safety department of your government, and not settled in civil court. These are criminal offenses, with optional jail holiday.
-
I would like to ask if it is possible to isolate each equipment using individual small UPS like this cheap one
http://www.apcbyquickserv.com/products/category/detail/index.php?idss=1900 (http://www.apcbyquickserv.com/products/category/detail/index.php?idss=1900)
and then operate each gears using offline mode base on UPS battery. Does this approach give a true isolation from main?
Not sure what true isolation is. Assuming the UPS's not connected to lines or ground, I would say that is truly isolated but once you connect the equipment to the line ... |O
Assuming a double conversion type and someone was stupid enough to float the ground, how are their transformers made? What sort of barrier do they have? Do they have a internal shield? My gut tells me these are not designed for it but I have never had a need to dig into it.
-
I would like to ask if it is possible to isolate each equipment using individual small UPS like this cheap one
http://www.apcbyquickserv.com/products/category/detail/index.php?idss=1900 (http://www.apcbyquickserv.com/products/category/detail/index.php?idss=1900)
and then operate each gears using offline mode base on UPS battery. Does this approach give a true isolation from main?
Not sure what true isolation is. Assuming the UPS's not connected to lines or ground, I would say that is truly isolated but once you connect the equipment to the line ... |O
Assuming a double conversion type and someone was stupid enough to float the ground, how are their transformers made? What sort of barrier do they have? Do they have a internal shield? My gut tells me these are not designed for it but I have never had a need to dig into it.
When the UPS is on battery power and not connected to mains (as some above have stated, you may have to initially connect the UPS to mains to get it started, then pull the plug from mains to get it to switch to battery), then it's isolated. The oscilloscope ground lead, internal UPS grounds, and the BNC ground shields are all at the same potential, floating.
This setup is truly isolated. There are no references from the oscilloscope to mains hot, neutral, or ground.
-
I would like to ask if it is possible to isolate each equipment using individual small UPS like this cheap one
http://www.apcbyquickserv.com/products/category/detail/index.php?idss=1900 (http://www.apcbyquickserv.com/products/category/detail/index.php?idss=1900)
and then operate each gears using offline mode base on UPS battery. Does this approach give a true isolation from main?
Not sure what true isolation is. Assuming the UPS's not connected to lines or ground, I would say that is truly isolated but once you connect the equipment to the line ... |O
Assuming a double conversion type and someone was stupid enough to float the ground, how are their transformers made? What sort of barrier do they have? Do they have a internal shield? My gut tells me these are not designed for it but I have never had a need to dig into it.
When the UPS is on battery power and not connected to mains (as some above have stated, you may have to initially connect the UPS to mains to get it started, then pull the plug from mains to get it to switch to battery), then it's isolated. The oscilloscope ground lead, internal UPS grounds, and the BNC ground shields are all at the same potential, floating.
This setup is truly isolated. There are no references from the oscilloscope to mains hot, neutral, or pground.
That, of course, implies a specific sequence of connection and disconnection.
It is easy to make a mistake, particularly when tired or distracted, and doubly so if more than one person is involved.
-
Yea, isolated until they reference the equipment back to the line they want to measure. Imagine having your scope's chassis riding on a 1200V bus. :-DD Maybe the signal generator gets referenced to another phase! :-DD Someone grabs the two. Seems like it could get interesting....
Probes don't cost all that much.
OP, curious why you don't just simulate the lines?
-
When applicable could one just use the UPS(unplugged from wall..running of it's battery) to POWER the *circuit* being tested....and leave the o-scopes ground and power plugged into the wall as usual.
-
I've noticed that several people around here use one of these things between their oscilloscope plug and the wall socket. I don't think it's safe at all.
(https://upload.wikimedia.org/wikipedia/commons/5/53/Cheater_plug_edited.jpg)
-
For isolation I use a battery powered scope which has two channels isolated from each other. For you I'd recommend diff probes. Assuming your department has ANY budget the cost shouldn't be an issue. If it's just you I'm sure you can afford them with such a fancy title.
-
I agree about the differential probes. By far the safest way to measure high voltages.
-
I don't think it's safe at all
Very unsafe, I'd say.
I have a special IEC lead for when I need to leave ground disconnected. On that, the ground lead is pulled back through the cable clamp of the plug so it is waving in the air outside the plug as a visual clue that things are not quite right (in addition to the hazard warning labels, etc).
-
I don't think it's safe at all.
Disconnecting the protective mains earth (PME) is indeed very dangerous. It kills scopes and people.
See, amongst many examples, https://www.eevblog.com/forum/testgear/ground-of-oscilloscope-always-connected-to-earth/msg1967150/#msg1967150 (https://www.eevblog.com/forum/testgear/ground-of-oscilloscope-always-connected-to-earth/msg1967150/#msg1967150)
-
Somewhere in this forum is an image of a rare case why you would float a scope. IIRC, it was a scope to monitor an RF transmitter powered by high voltage. To protect the engineers the scope is behind an acrylic glass and the few controls needed are extended by plastic rods to provide sufficient distance to the high voltage.
-
Somewhere in this forum is an image of a rare case why you would float a scope. IIRC, it was a scope to monitor an RF transmitter powered by high voltage. To protect the engineers the scope is behind an acrylic glass and the few controls needed are extended by plastic rods to provide sufficient distance to the high voltage.
The scope below was sold on fleabay, with no indication of why it was floated. It remains a good illustration of what is necessary to float a scope "properly" - and hence why you shouldn't :)
Fundamentally the right way to achieve the measurement is almost always to use the right probe (isolating and/or HT and/or differential) or an special-purpose scope with isolated inputs.
(http://bristol.hackspace.org.uk/wiki/lib/exe/fetch.php?media=oscilloscope:tektronix-howtofloatascope.jpg)
-
Floating the scope only works if there is no Y capacitor inside, but a real foiled transformer. When there is a Y-cap, you'll spark your board to death.
*learned from experience* :palm:
-
Let's imagine that one can actually "float" the scope, by whatever means: isolation transformer, a disconnected UPS, or whatever. One must remember that all BNC shells are now not at safe ground potential, but at whatever potential they are connected to in circuit. Remember too that they are all connected together; you can't just arbitrarily probe four signals by connecting the four ground clips to four separate points.
Here is another huge consideration: once you connect any/all of the "floating" ground clips to an arbitrary point in the circuit, all of the probe cables' ground leads and shields, as well as the chassis of the scope are (obviously) connected to that node. That's a big antenna, and a big conductor to provide capacitance to actual ground (i.e. to most other conductive things in the area). In other words, you will inject a large amount of RF into that node, and you will load that node with lots of pF to ground, much more than any proper differential probe would do. Even a correct probing technique can load a circuit to the point that the measurement can't be taken as 'truth' without consideration of the loading. If anyone thinks that the act of connecting a "floating" ground clip to an arbitrary node will not significantly affect the circuit, he is a fool.
-
Let's imagine that one can actually "float" the scope, by whatever means: isolation transformer, a disconnected UPS, or whatever. One must remember that all BNC shells are now not at safe ground potential, but at whatever potential they are connected to in circuit. Remember too that they are all connected together; you can't just arbitrarily probe four signals by connecting the four ground clips to four separate points.
Here is another huge consideration: once you connect any/all of the "floating" ground clips to an arbitrary point in the circuit, all of the probe cables' ground leads and shields, as well as the chassis of the scope are (obviously) connected to that node. That's a big antenna, and a big conductor to provide capacitance to actual ground (i.e. to most other conductive things in the area). In other words, you will inject a large amount of RF into that node, and you will load that node with lots of pF to ground, much more than any proper differential probe would do. Even a correct probing technique can load a circuit to the point that the measurement can't be taken as 'truth' without consideration of the loading. If anyone thinks that the act of connecting a "floating" ground clip to an arbitrary node will not significantly affect the circuit, he is a fool.
And in some cases (e.g. with FETs) such extra capacitance can damage the UUT.
-
I would not really consider using anything but an aproved isolation transformer. These are not very expensive and have a well proven useage case. There is nothing wrong with being inventive but I am very very lazy.
-
I would not really consider using anything but an aproved isolation transformer. These are not very expensive and have a well proven useage case. There is nothing wrong with being inventive but I am very very lazy.
Thinking you are safe because you are using an isolation transformer is as accurate as thinking you are safe because you have a network firewall.
Botjh firewalls and isolation transformers have many many interesting and subtle traps for the unwary. The Dunning-Krueger syndrome is a real issue.